1. First, do no harm

The work of humanitarian design is high stakes. We’re not talking about making it easier to share jokes and photos or to show friends how sweet our lives are. By our definition humanitarian design is concerned with weighty issues like suffering and injustice, so what we do really matters and we have a responsibility as large as the scope of our work. As many people as we can help, we can hurt.

An example comes from the work of Google’s Crisis Response team during Hurricane Sandy. One of the key needs of people on the ground was to find gas stations that actually had gas. The team found a data feed directly from one of the oil companies and surfaced it on a map of the area. The only trouble was that, as we later found out, the feed was only refreshed every few hours. That means we were potentially encouraging people to drive across town only to find an empty station. Sometimes bad data is worse than no data.

Doctors have been doing humanitarian work for a long time and they have this fundamental principle: “first, do no harm.”

This doesn’t mean we can never do anything risky that may possibly cause harm but it’s all too easy to make things worse not better, so when considering an intervention we should always look for and carefully weigh the possible ill effects against the potential benefits.

2. Think big, build small

Unfortunately doing no harm is especially difficult in humanitarian spaces because of the types of problems we’re trying to solve.

Designers often think in terms of individual products or, at best, cohesive services, but things like poverty, ineffective governance, crime, and, yes, even tea involve extremely complex, interdependent systems.

And there are endless examples of the kinds of unintended consequences that can arise from good intentions applied to complex systems. As just one simple one, while irrigation schemes provide people with water for agriculture, they can increase waterborne diseases that have devastating health effects.

So how well do we need to understand all the connections? How do we deal with adjacent areas? What approaches to designing within complex systems work best?

Here’s a second principle: think big, build small. This is the equivalent of the old bumper sticker “Think globally, act locally,” the need to keep the system in mind and understand as many of the relationships as possible without being afraid to solve a specific problem in that context.

This is the way things happen at Google (where I work). Can you imagine if one team tried to build all of Google as it exists today?

The encouraging corollary to the problem of unintended consequences is that small nudges can lead to big changes. Buckminster Fuller explained this with the analogy of the trim tab, a small piece connected to the end of a rudder (or elevator in an airplane). Moving this small piece can adjust the neutral position of the rudder, changing the direction of the entire vessel.

The little individual can be a trim tab. Society thinks it’s going right by you, that it’s left you altogether. But if you’re doing dynamic things mentally, the fact is that you can just put your foot out like that and the whole big ship of state is going to go. — Buckminster Fuller

3. Be humbly ambitious

It’s an uncomfortable and therefore little-discussed fact of humanitarian design that many of us are in part motivated by ego gratification. We get a thrill from working in extreme conditions, we earn social capital, we feel good about doing good. I have no fundamental problem with this insofar as it motivates designers, but it becomes a real problem when the design decisions themselves are motivated by designers’ egos, blinding them to real needs and conditions.

Here’s an example:

Water availability and cleanliness is one of the biggest humanitarian issues we face. 1 in 10 people does not have adequate access to safe water and every day more than 2000 children die of diarrhea caused by unsafe water and poor sanitation.

Along with great need, there’s something pure about this as a problem space. What could be more fundamental to life than water? As a result, there are dozens of different designs for water pumps and wells intended to make it easier for folks living in rural places to get water. The diversity of ideas is great for innovation, but means there’s lots of competition for resources. And because it’s incredibly difficult to measure the impact of such things, the most effective designs don’t necessarily get the funding.

Playpumps are a classic example.

This sounds like such a great design success story: the idea is that by spinning on a merry-go-round connected to a water pump, children could generate plentiful, clean water without the time-consuming labor of traditional hand pumps. Kids get a place to play, communities get drinking water, and girls and women — who traditionally do much of the hard work of getting water for their families — have more time for things like attending school or starting a small business. Plus billboards on the water tanks would bring in advertising revenue to help fund the pumps’ maintenance and provide public health messaging. Brilliant.

Not surprisingly, the pumps generated a ton of positive publicity (think Jay-Z benefit concerts) which led directly to major funding — $16 million from USAID, for example.

The only problem is that the idea didn’t work.

The pumps are expensive and difficult to repair. It takes about a day’s worth of playing to fill a tank. You’d need an army of children with tons of free time on their hands to get the job done. There’s little money to pay for advertising in rural Sub-Saharan Africa. And they’re harder to pump than many existing designs. So what we end up with is embarrassed women struggling to turn $14,000 merry-go-rounds.

But they’re photogenic and an appealingly clever-sounding idea so they got a lot of attention and funding.

By contrast, the AfriDev is a more-traditional lever-operated hand pump. It’s not so photogenic but it handily outperforms the PlayPump and is a public domain design. It costs $500ish. To me, that’s brilliant design.

So by all means let’s harness hubris to motivate hard work and ambition but let’s also be wary of how that motivation impacts the decisions we make.

4. Design for dignity

I’ve been talking a lot about systems. But ultimately this work is about people, and people have not only needs but desires and feelings and complex psychology.

What does it mean to people to be “impoverished” or “in need” and how does that change what and how we might design positive interventions for them?

There’s an amazing essay called “Glory” by a Kenyan author named Binyavanga Wainaina which addresses this in the context of supplying energy to rural populations. An excerpt:

I was twelve years old, in a small public school in Nakuru [Kenya].

One day, the whole school was called out of class. Some very blond and very serious people from Sweden had arrived. We were led to the round patch of grass next to the parade ground in front of the school, where the flag was. Next to the flag were two giant drums of cow shit and metal pipes and other unfamiliar accessories. We stood around, heard some burping sounds, and behold, there was light.

This is biogas, the Swedes told us. A fecal matter. It looks like shit — it is shit — but it has given up its gas for you. With this new fuel you can light your bulbs and cook your food. You will become balanced dieted; if you are industrious perhaps you can run a small biogas powered posho mill and engage in income generating activities. We went back to class. Very excited. Heretofore our teachers had threatened us with straightforward visions of failure. Boys would end up shining shoes; girls would end up pregnant.

Now there was a worse thing to be: a user of biogas.

Dignity is a fundamental human right — the very thing we’re addressing. But it’s also a critical design constraint when working in humanitarian areas. Humanitarian designers design for dignity. Products and services, no matter how brilliant, can only have impact if people actually use them. And people are going to resist adopting things that make them feel inferior.

5. Design for ownership

This idea of the criticality of adoption — that a product can only have positive impact if people adopt and use it — is about how a technology is introduced.

But that moment of introduction is not only a question of failure or success. It can alter the very nature of the intervention being made. Here’s an example from when Gilberto Gil was Minister of Culture for Brazil (how cool is that?!).

When Gil & Claudio Prado set up Pontos de Cultura (Cultural Hotspots) in previously unconnected areas they thought deeply about how to introduce people to information and communication technologies. Each spot got a recycled computer as well as a video camera and microphones. There are two ideas here. First to teach people that computers are repairable and therefore belong to them and second to introduce technology as a tool for self-expression as opposed to a conduit for passive consumption.

Designing for people to help themselves rather than being dependent on external aid is the only reasonable path to sustainable interventions in a system. Things break and need to be fixed. Conditions change so tools need to evolve.

This is behind the idea of “appropriate technology” — making things that can be locally maintained and modified by the people using them. But there’s more to it. If we design so that people own the means of cultural production they become full digital citizens and the technology becomes a tool which they can use to change the world around them in ways that serve their needs.

6. Measure impact not use

There’s a lot here about figuring out what to make, but what about checking to see if it works?

As our ability to measure the world has exploded, we’ve become increasingly infatuated with measurement. And that’s great to the extent that it leads to decisions which are grounded in reality rather than assumptions, conjecture, and hope. But we can only measure certain things and there’s always a danger that we’ll design to what we can measure.

Unfortunately, compared to, say, running an ecommerce web site, it’s really hard to measure social outcomes because the changes we’re striving for are typically offline. We need to know whether people who use a product are more healthy, vote more, are better able to conserve energy, and so on than those who don’t.

The most typical approach for social impact software is to use proxies for impact (such as usage) with the assumption that we’ve done a good job and if more people use the product then more good will be done. This can work if we also measure the validity of the proxies themselves. But it will always be somewhat tenuous.

There are effective established techniques— like randomized controlled trials — but they can be costly, time-consuming and take a lot of expertise. They’re also hard to scale because what works in one place may not work in another. And it’s hard to iterate rapidly when each round of measurement takes months or years to complete.

There are also precedents in political advertising and public health campaigns, using surveys to correlate (self-reported) behavior with specific messages. And online traces can increasingly be correlated algorithmically with offline behavior (creepy as that may be).

None of these techniques is perfect, but they all have potential to bridge the online-offline gap. And just being aware of the need reminds us to step back from the numbers and incorporate other ways of evaluating success.

7. Design with, not just for

Remember Playpumps? A large part of the problem was that they were designed from outside and then imposed on the people they were intended to help.

Here’s a story with a very different outcome. In the early 1990s, a group of researchers was looking for a sustainable way to improve nutrition in rural Vietnam. They knew the solution would need to come from within local communities and were influenced by the theory of “positive deviance.”

Positive Deviance is based on the observation that in every community there are certain individuals or groups whose uncommon behaviors and strategies enable them to find better solutions to problems than their peers, while having access to the same resources and facing similar or worse challenges. — The Positive Deviance Initiative (http://www.positivedeviance.org/)

Based on this idea they worked with community members to find “well nourished children from poor families (the positive deviants) in [the] most vulnerable populations with [a] high degree of childhood malnutrition.”

They found that — unlike their peers — these families

• fed their children many meals throughout the day
• ground up tiny shrimps and snails they found in the fields, which others considered unhealthy.

They then worked with the community to incorporate these behaviors into the lives of others.

This last principle is fundamental. It’s part of the answer to every challenge I’ve mentioned: design with, not just for.

People may not always have the expertise we bring as designers or technologists, but nor will we have their point of view (unless we’re designing for ourselves). People know their situations deeply.

Rather than being a burdensome but necessary part of the design process this is a fantastic opportunity. If we involve people in solving their own problems — either through thoughtful research or participatory design — we gain access to insights and opportunities which would otherwise remain hidden. And that means a much greater chance those solutions will resonate and be used sustainably and impactfully by others.